1. Introduction

Sexual and gender minorities are at higher risk for problems related to alcohol and drug use compared with heterosexuals [1–3]. Data show that sexual minority adults are more than twice as likely as their heterosexual counterparts to use illicit drugs and almost twice as likely to suffer from a substance abuse disorder [4, 5]. For example, opioid use, including abuse in the past year, was 9% among sexual minority adults compared to 3.8% among the overall adult population [2]. Alcohol and drug use are associated with several adverse health outcomes and comorbidities, including HIV, overdose, mortality, and other sexually transmitted infections [6–8]. Sexual minorities disproportionately suffer from adverse effects of alcohol and drug use and are more likely to have additional comorbidities, including HIV and mental health disorders [9, 10]. BSMM experience more significant risks for these comorbidities, including HIV and STIs, compared with white SMM despite reporting less frequent sexual risk behaviors [11, 12]. The reasons for the disproportionate risks to these conditions among BSMM are due intersecting structural adversities, including social stressors [13]. Understanding factors associated with drug and alcohol use problems among Black SMM is critical, given that this group disproportionately suffer from adverse health effects.

According to the minority stress theory, sexual minority individuals experience multiple stressors, including enacted stigma, prejudice, identity concealment, internalized stigma, and anticipated stigma [13]. As a result of these stressors, some may resort to substance use as a coping mechanism [14]. Prior studies have identified associations between minority stress and increased likelihood of alcohol and drug use among sexual minorities [13, 15, 16]. Also, existing data show that Black SMM are disproportionately affected by minority stressors [6, 17–19]. Black SMM may experience discrimination due to their sexual identity and/or their race/ethnicity. Some studies show that they may use substances such as heroin, cocaine, and methamphetamine to cope with social stress [14, 20]. Black SMM often experience concomitant challenges from poverty, stigma, violence, and high community HIV and STI prevalence that produce a cumulative effect on substance use, sexual risks, and subsequent HIV infection [14, 18]. Several psychosocial stressors are associated with substance use among Black SMM. These include depression, internalized homophobia, and trauma [21–24]. Roots of psychosocial problems among Black SMM have been shown to lie in early experiences of stress, stigma and discrimination, which in turn increase the likelihood of mental health problems, substance abuse, and sexual risk behavior [25].

Childhood sexual abuse (CSA) is among the early childhood stressors reported among Black SMM [18, 26]. CSA has been linked to HIV risk, mental health problems, substance abuse, and intimate partner violence among many populations [27–29]. CSA has garnered attention as a precursor to increased risk for HIV, especially given meta-analytic evidence of elevated rates of CSA among MSM [30]. However, the majority of these studies have focused almost exclusively on White SMM. Limited studies have documented CSA among Black SMM. Notable exceptions include a study of HIV-positive Black SMM in New York City which found CSA to be a precursor to HIV infection, substance misuse, and intimate partner violence [31]. To build on this literature, we aimed to assess associations of CSA, alcohol use, and drug use problems in a multi-site study of adult Black SMM.

2. Materials and methods

2.2. Study sample

From a total of 1553 men at baseline, we excluded 438 with missing data on the independent (CSA) and dependent variables (drug and alcohol use). We also excluded 99 participants with missing data on covariates (depression), and this resulted in the analytical sample of 1016 men with complete data (Fig 1).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Analytic sample.

https://doi.org/10.1371/journal.pone.0279238.g001

2.4. Data analysis

This analysis is restricted to participants with complete data on the dependent variables (6-month drug use and alcohol use problems), the independent variable (CSA), and covariates. We used frequencies and percentages to describe the study sample (Table 1). Also, we conducted descriptive statistics by CSA for all the study variables to identify any significant differences between those who reported history of CSA and those with no history of CSA. Chi-square tests were performed for all categorical variables (Table 2). Finally, we conducted hierarchical logistic regression analyses to determine whether CSA was associated with: (1) 6-month drug use and (2) alcohol use problems, adjusting for covariates. We entered CSA status at the first step (model 1), followed by socio-demographic characteristics (model 2), and lastly, depressive symptoms (model 3). The goal was to examine whether the strength of the association between the independent variable (CSA) and the dependent variables (drug use and alcohol use problems) was weakened with further additions of other variables in the model. Statistical analysis was conducted using Stata Statistical Software Release 14 (College Station, TX).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Overall characteristics of the study sample (N = 1016).

https://doi.org/10.1371/journal.pone.0279238.t001

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Descriptive summary of study variables by childhood sexual abuse (N = 1016).

https://doi.org/10.1371/journal.pone.0279238.t002

3. Results

Table 1 presents the characteristics of the sample (N = 1016). More than half (53.5%) of the study participants were between the ages of 36 and 55 years with a very small percentage of older than 55 years (6%). Almost 50% had ever used at least one of the following drugs: poppers, crack cocaine, powder cocaine, methamphetamine, heroin, unprescribed Vicodin or ketamine, and other non-prescribed drugs in the last six months and close to 30% had a problem with alcohol use in the past six months. Overall, most participants reported experiences of childhood sexual abuse (70%). Also, 43% of the sample reported depressive symptoms.

Drug use during the past six months and alcohol use problems differed by CSA history (refer to Table 2). Participants who had ever experienced any CSA reported higher drug use during the past six months and alcohol use problems compared to those without a history of CSA (p<0.001). Participants with a history of CSA were significantly more likely to report depressive symptoms than those without a history of CSA (p<0.001). We also observed that participants with a history of CSA were significantly more likely to report insufficient income in the past six months than those with no CSA history (p = 0.02). Additionally, participants with a CSA history were more likely to have provided or received sex for money or drugs in the past six months than those who were not abused (p<0.001).

Tables 3 and 4 show the results of the hierarchical logistic regression analysis. In Table 3, experiences of CSA in model 1 is associated with higher odds of drug use in the past six months compared to the non-CSA group (odds ratio (OR) = 1.71, 95% CI 1.30, 2.24). We included socio-demographic factors (age, sexual identity, education, employment, income, and living conditions with CSA variable in model 2, and found that participants aged 36–55 years and those above 55 years had higher odds of drug use than participants aged 18–35 years, (OR = 2.16, 95% CI 1.64, 2.85) and (OR = 2.34, 95% CI 1.42, 4.16) respectively. Participants who reported insufficient money in the past six months had increased odds of drug use (OR = 1.45, 95% CI 1.11, 1.89). The depressive symptoms variable was added in the final model 3 and was significantly associated with drug use (OR = 1.42, 95% CI = 1.00, 1.70; p = 0.05). The relationship between CSA and drug use remained significant. Also, age and insufficient money for rent in the past six months remained significantly associated with drug use during the past six months.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Variables associated with drug use: Results from hierarchical logistic regression models (N = 1016).

https://doi.org/10.1371/journal.pone.0279238.t003

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Variables associated with alcohol use problems: Results from hierarchical logistic regression models (N = 1016).

https://doi.org/10.1371/journal.pone.0279238.t004

In Table 4, experiences of CSA in model 1 was associated with 74% increased odds of alcohol use problems. We included socio-demographic factors with CSA variable in model 2. The strength of their association with alcohol use remained significant (p<0.001). In addition, insufficient income in the past six months remained significantly associated with alcohol use problems (OR = 1.66, 95% CI 1.23, 2.24). The depressive symptoms variable was added in the final model 3 and was significantly associated with alcohol use problems (OR = 2.25, 95% CI 1.69, 3.00). The relationship between CSA and alcohol use problem remained significant (OR = 1.52, 95% CI 1.09, 2.12). Participants who reported not enough money for rent in the last six months had 56% increased odds of alcohol use problems (OR = 1.56, 95% CI 1.15, 2.11).

4. Discussion

Findings from this six-city study of Black sexual minority men reveal a high prevalence of CSA which was associated with alcohol and drug use problems in the study sample. 70% of the sample had experienced CSA. Overall, 29% and 49% of BSMM reported alcohol use and drug use problems, respectively, and CSA was significantly associated with both risk behaviors independent of sociodemographic and depressive symptoms confounding variables. Our data suggest that interventions to address alcohol and drug use problems among Black SMM should address the role of early life adversities, including the potential role of CSA, as a significant factor contributing to substance use. These findings are particularly relevant in light of national efforts to reduce disparities in substance use among African Americans and to provide them with culturally-relevant intervention and treatment programs [37].

Our data align with results from a study by Wu (2018) which found 28% CSA prevalence in a New York City sample of 1,002 Black SMM. In that study, CSA history was associated with 50% greater odds of both binge drinking and illicit substance use, which are in line with our findings. However, the prevalence of any CSA was substantially higher in our sample (70%) compared to findings reported in the Wu study, which might reflect a crucial difference in study inclusion criteria. Our study included Black SMM who reported condomless sex during the past six months whereas the study by Wu included Black SMM who reported any sex in the past three months regardless of condom use or nonuse. Indeed, previous research has indicated that CSA is associated with higher rates of condomless sex and HIV risk [29], potentially accounting for the high CSA prevalence in our sample. Moreover, in a previous systematic review and meta-analysis of CSA-related risk factors among SMM [30], CSA was associated with an 85% greater odds of recent condomless sex, 54% greater odds of HIV-positive status, as well as being correlated with substance use in seven of eight published studies (meta-analysis of CSA-substance use association was not conducted due to heterogeneity in measurement). Taken together, the current study provides further evidence for a syndemic of CSA, substance use, and HIV risk among Black SMM that warrants consideration for the implementation of interventions to reduce health disparities in this population.

Notably, our findings also bring attention to the role of depression and financial hardship as potential components of the syndemic comprising CSA, substance use, and HIV risk among Black SMM [18]. With regard to our primary research question (on the association between CSA and substance use behaviors), depression and financial hardship variables were considered confounders and thus were adjusted for in regression analyses. However, depression and financial hardship remained significantly associated with CSA in the fully adjusted models, and were each correlated with alcohol use problems and drug use. These persistent associations highlight a need for nuanced, multi-level conceptual models to guide public health interventions that address substance use and HIV risk among Black SMM. In this case, substance use and HIV behavioral risk factors among Black SMM should be understood in the context of economic hardship and psychosocial vulnerability in this group, as well as through the lens of a life course perspective that includes the possibility of CSA or other forms of early-life abuse [14]. Indeed, studies that have integrated the treatment of CSA trauma-related symptoms with HIV risk reduction counseling for SMM have observed HIV risk reduction among SMM [38].

Limitations to this research must be considered. First, as noted, inclusion criteria stipulated that Black SMM participants must have engaged in condomless sex during the past six months, which might have resulted in the high CSA prevalence observed in the sample. Second, self-report data might have been influenced by social desirability bias. In particular, due to associated stigmas and trauma, reports of CSA are especially prone to challenges in disclosure and recall. At the same time, ACASI was used with participants to minimize these effects. Third, the use of cross-sectional data prevents causal inferences, although the timing of the CSA variable temporally precedes all other risk variables reported here. Fifth, the study sites included larger metropolitan areas with high HIV prevalence, thereby limiting generalizability.

In conclusion, these findings underscore the need to consider CSA as a factor in the design and delivery of prevention intervention and treatment programs for alcohol use problems and drug use with Black SMM. Findings from this study suggest a need for more research to more rigorously measure the psychosocial and life course factors that may account for the link between CSA and substance use behaviors among Black SMM, including depression, trauma, as well as economic hardship. Moreover, findings provide further evidence for a syndemic between CSA, substance use, and HIV risk among Black SMM. These findings also suggest need to understand the role of adverse childhood experiences on health outcomes for BSMM. Future studies should examine the role of intersecting social determinants of health and health outcomes of stigmatized and vulnerable populations like BSMM. Also, based on these findings, intersectional inequalities negatively affect health outcomes of stigmatized populations, thus, there is need for interventions that address social and structural factors to reduce health inequities experienced by this population. Also, future interventions should be designed with meaningful input from the BSMM community, to understand the complexity related to these health challenges and to minimize the possibility of stigmatization due to the sensitive nature of CSA.

Acknowledgments

We thank the HPTN project staff for their support in providing data for this manuscript. We also wish to acknowledge study participants for their participation in HPTN061 project.